01 — Evidence
Evidence Summary
EGCG's evidence base is broad rather than deep: human data anchors mainly to pharmacokinetics and one hematologic-cancer trial, animal corroboration exists but is limited to a handful of independent findings, and cell-model data is the most extensive layer by far. Reading across all three tiers together gives the fairest picture.
Human
Clinical Record
Blood cancer trial; mixed prostate data
A Phase 2 trial in early-stage chronic lymphocytic leukemia reported sustained reductions in circulating leukemia cells and swollen lymph nodes in a meaningful share of patients over six months of daily dosing — real biological activity, not a cure.
- Fasted dosing substantially increases how much EGCG reaches the bloodstream compared with dosing alongside food
- Prostate cancer chemoprevention trials are mixed — one reported fewer cancer diagnoses, a larger later trial found no difference on that same measure
- No human trial has yet shown EGCG shrinks an existing solid tumour
Animal
Preclinical Signal
Xenograft and chemoprevention models
In a colon-cancer xenograft model, EGCG was associated with a markedly smaller tumour, fewer new blood vessels, and more tumour-cell death, tied to suppressed VEGF signalling.
- A separate study confirmed fat-synthesis blockade triggers tumour-cell death in both cell and animal colorectal models
- A chemically stabilised version of EGCG outperformed native EGCG in breast and prostate xenograft models
- In mice, EGCG reduced diet-induced colon precancerous lesions and separately reduced liver injury markers in an induced liver-injury model
In Vitro
Cell Model Data
Deepest layer; concentrated in colon cells
Cell-model data is EGCG's deepest evidentiary layer, converging most consistently in colorectal cancer cells across receptor blockade, growth-pathway suppression, metabolic pressure, and programmed cell death.
- EGFR/HER2 receptor blockade reported across five colon-cancer cell lines in one study
- Wnt/β-catenin suppression linked to both slowed growth and reduced stem-like cell survival
- Two independent programmed-cell-death mechanisms confirmed through receptor- and enzyme-level testing
Human
Clinical Record
The clearest human signal for EGCG comes from blood cancer, not a solid tumour. In a Phase 2 trial of Polyphenon E — a standardised green tea extract — standardised to 2,000 mg EGCG per dose, twice daily, in 42 patients with early-stage chronic lymphocytic leukemia, a sustained reduction in circulating leukemia cells was reported in roughly a third of patients, and a meaningful reduction in swollen lymph nodes in more than two-thirds, over six months of daily dosing.[4] This is real biological activity against measurable disease burden — not a cure, and the trial did not establish delayed disease progression, reduced need for later treatment, or a survival benefit. How much EGCG actually reaches the bloodstream from a given dose depends heavily on whether it's taken with food: a randomised trial found that fasted dosing substantially increased systemic exposure compared with dosing alongside a light meal — detailed further under Pharmacokinetics and Administration below.[2]
Continue reading — full research detail+
Chemoprevention is where the human record gets genuinely mixed, and both trials used mixed green tea catechin preparations rather than isolated pure EGCG, so results cannot be attributed to EGCG alone. A randomised, placebo-controlled proof-of-principle trial in men with high-grade prostatic intraepithelial neoplasia (HGPIN) — a precursor lesion, not cancer itself — found significantly fewer prostate cancer diagnoses after one year of daily green tea catechins versus placebo.[24] A larger, later randomised trial testing a similar regimen found no significant reduction in its primary endpoint of one-year prostate cancer incidence; a secondary analysis combining cancer diagnoses with atypical small acinar proliferation (a related precursor finding) favoured the treatment arm, but that analysis was exploratory and does not overturn the negative primary result.[25] Both trials are real; neither is allowed to stand alone as the answer here.
Signal maturity: human research currently establishes a real, if modest, biological-activity signal against measurable disease burden in one blood cancer, a well-characterised absorption profile, and a genuinely unresolved chemoprevention question in prostate tissue. No human trial has yet demonstrated that EGCG shrinks an existing solid tumour or improves survival.
Animal
Preclinical Signal
The strongest single piece of animal evidence for EGCG is a colon-cancer xenograft study: in mice carrying human colon tumour grafts, EGCG treatment was associated with a markedly smaller tumour, fewer new blood vessels feeding it, and more tumour- and vessel-cell death than untreated controls, tied mechanistically to suppressed VEGF signalling measured in the same tumours — one of the few findings in this profile with matched cell-level mechanism and living-tumour outcome data from the same study.[3] A second, independent study reported that EGCG triggers tumour-cell death by shutting down fat-synthesis machinery in colorectal cancer cells, and confirmed this in a matched animal model as well as in the dish.[17]
Continue reading — full research detail+
A chemically stabilised version of EGCG (a peracetate prodrug, designed to resist the compound's own rapid breakdown) outperformed native EGCG on tumour growth in both a breast-cancer and a prostate-cancer xenograft model, in each case alongside reduced tumour blood-vessel growth and more tumour-cell death — detailed further under Pharmacokinetics and Administration's Formulation section below.[26,27]
Separately from tumour-suppression evidence, two animal findings speak to EGCG's effect on the host rather than the tumour. In mice, EGCG reduced colon precancerous lesions induced by a dietary carcinogen, associated with activation of a cellular antioxidant defence pathway (Nrf2) — a real chemoprevention signal in a living animal, not just a cell dish.[12] In a separate mouse model of induced liver injury, EGCG reduced liver scarring, oxidative stress, and inflammation — a genuine host-protective finding, addressed in full, alongside a real tension with EGCG's own hepatotoxicity risk, under Pathway Interaction Profile and Safety below.[8]
Signal maturity: animal models provide EGCG's clearest tumour-suppression corroboration beyond cell-dish data, but that corroboration is limited to a handful of independent studies rather than a broad, repeated pattern across many tumour types and research groups — and no animal finding here has yet been confirmed in a human oncology trial.
In Vitro
Cell Model Data
Cell-model data is EGCG's deepest evidentiary layer by a wide margin, and colorectal cancer cells specifically are where its mechanisms converge most consistently — receptor blockade, growth-pathway suppression, metabolic pressure, and programmed cell death have all been reported across the same tumour type, sometimes in the same cell lines.[1,5,6,15,16,20] EGCG has been reported to block activation of the EGFR/HER2 receptor pair in five colon-cancer cell lines in one study, preferentially slowing the cancer cells over normal colon cells from the same tissue.[1] A separate, independently replicated mechanism — suppression of the Wnt/β-catenin growth pathway — has been reported both to slow colon cancer cell growth directly and to block the stem-like cell state colorectal tumours rely on to survive treatment and reactivate.[5,6]
Continue reading — full research detail+
EGCG has also been reported to apply metabolic pressure on colon cancer cells specifically, activating an energy-sensing enzyme (AMPK) that in turn reduces the cell's glucose-uptake machinery and a key inflammatory signal (COX-2) — three separate metabolic and inflammatory effects traced to one upstream switch.[15] The same AMPK activation was independently reported to suppress a central growth-survival pathway (PI3K-Akt-mTOR) in the same cell line.[16] A cellular stress-response pathway (the unfolded protein response) has also been reported to trigger programmed cell death in colorectal cancer cells exposed to EGCG.[20]
Two distinct cell-death mechanisms have been reported in colon cancer cells specifically: EGCG combined with a death-receptor ligand (TRAIL) triggered programmed cell death confirmed to depend on a specific surface receptor (DR5) — removing that receptor blocked the effect entirely, a genuine mechanistic confirmation of a combination effect, not EGCG killing tumour cells alone.[7] Outside colorectal tissue, EGCG has been reported to block a growth-signalling protein (STAT3) in pancreatic cancer cells, reducing their ability to migrate and invade,[9] and to block a distinct growth-signalling pathway (TGF-β/Smad) in thyroid cancer cells, reversing the cellular transition those cells use to invade and spread.[21] A metabolic enzyme EGCG is known to inhibit (glutamate dehydrogenase, acting one step downstream of where glutamine enters the cell) has separately been reported to sensitise glioblastoma cells specifically when glucose metabolism or a growth-survival pathway (Akt) was already impaired — a conditional vulnerability, not a standalone glutamine-starvation effect — and, combined with a second compound (fenretinide), to trigger cell death in a high-risk childhood cancer (MYCN-amplified neuroblastoma) cell model.[22,23]
Signal maturity: cell-model research offers the deepest and most consistent mechanistic detail of any evidence tier for EGCG, converging especially strongly on colorectal cancer cells — but the concentrations used to produce several of these effects exceed what oral dosing has been shown to achieve in human plasma, detailed under Pharmacokinetics and Administration's Concentration Gap section below, and mechanism alone does not confirm real-world benefit.
Advertisement
02 — Pathways
Pathway Interaction Profile
EGCG engages several distinct biological pathways relevant to tumor behavior, grouped below by the functional role each one supports. This includes direct anti-tumour mechanisms and, further down, a separate set of pathways supporting the body's own resilience.
EGCG's Contain classification rests on reported suppression of tumour blood-vessel growth, an inflammatory signal that primes a tumour's surroundings, the cellular transition tumour cells use to detach and invade, and the stem-like state that lets dormant cancer cells survive treatment and reactivate.
Block Seeding & Niche Formation
Research concerning formation of supportive pre-metastatic niches at distant sites.
Angiogenesis / VEGF / HIF-1α
EGCG has been reported to suppress VEGF-driven blood-vessel growth in colon cancer cells, and in a matched xenograft model reduced tumour volume, new-vessel density, and tumour-cell proliferation while increasing both tumour- and vessel-cell death — one of the strongest single findings in this profile, with matched cell-level mechanism and living-tumour outcome data from the same study.[3]
COX-2 / PGE₂
EGCG has been reported to suppress a key inflammatory signal (COX-2/PGE₂) in colon cancer cells through activation of an energy-sensing enzyme (AMPK); directly blocking that enzyme reversed the effect, confirming AMPK as the driver rather than a coincidental correlate.[15]
Prevent Tumour Cell Shedding
Research concerning invasion and escape from existing lesions (EMT and ECM breach).
EMT & metastatic invasion
In thyroid cancer cells, EGCG has been reported to block a growth-signalling pathway (TGF-β/Smad) that drives the cellular transition tumour cells use to detach and invade, reversing markers of that transition and directly reducing invasion and migration.[21]
Prevent Dormant Reactivation
Research concerning wake-up signalling and reactivation of dormant disseminated tumour cells.
Cancer stemness (CD44, ALDH, Nanog/Sox2)
EGCG has been reported to block the stem-like state colorectal cancer cells rely on to survive treatment and reactivate, suppressing spheroid formation and stem-cell marker expression through the same Wnt/β-catenin pathway behind its growth-suppressing effect — and restoring that pathway's activity reversed the anti-stemness effect, confirming the mechanism directly.[6]
EGCG's Starve classification is anchored in reported disruption of how tumour cells fuel themselves — the glucose-uptake machinery cells rely on for fast energy, the fat-synthesis pathway cells use to build new membrane, and a metabolic enzyme sitting one step downstream of glutamine uptake that tumour cells depend on to feed their own growth.
Glucose Axis Pressure
Research concerning glycolytic ATP production and glycolytic intermediates used by cancer cells.
Aerobic glycolysis (Warburg effect)
EGCG-driven AMPK activation has been reported to reduce the glucose transporter GLUT-1 in colon cancer cells, alongside the same COX-2 and VEGF suppression described under Contain above — three separate effects traced to one upstream switch. The cited study measured GLUT-1, COX-2, and VEGF expression specifically, not glucose uptake, lactate production, or extracellular acidification directly — so this is best read as glucose-uptake pressure rather than a full measured Warburg-effect reversal.[15]
Lipid Axis Pressure
Research concerning membrane synthesis and lipid-driven signalling capacity.
Lipogenesis
EGCG has been reported to trigger programmed cell death in colorectal cancer cells by shutting down the fat-synthesis machinery those cells depend on to build new membrane — tested in both cell and animal models, making this the clearest standalone evidence of EGCG acting as a Starve-pathway compound. The in vitro concentrations used to establish this (IC50 500–800 µM) were substantially higher than in most other findings on this page — roughly 70- to 110-fold above the highest concentration confirmed in human plasma, detailed under the Concentration Gap section below — a real caveat on an otherwise animal-corroborated finding.[17]
Amino Acid / Protein Access Pressure
Research concerning nitrogen availability, amino-acid access, and biomass synthesis.
Glutaminolysis
EGCG inhibits an enzyme (glutamate dehydrogenase) that sits one step downstream of where tumour cells take in glutamine, rather than blocking that uptake directly. This downstream interference has been reported to sensitise glioblastoma cells specifically when their glucose metabolism or Akt signalling was already impaired — a conditional vulnerability rather than a standalone glutamine-starvation effect — and, in combination with a second compound (fenretinide), to trigger cell death in MYCN-amplified neuroblastoma cells. Broad suppression of tumour glutaminolysis has not been demonstrated in humans.[22,23]
EGCG's Weaken classification reflects reported attrition of the signalling tumour cells depend on to keep growing and manage internal stress — receptor-level growth signalling, the central survival pathway tumours co-opt to keep proliferating, and a cellular stress-response cascade that tips toward programmed cell death under sustained pressure. Human clinical evidence sits alongside these cell-model mechanisms rather than confirming any specific one of them: in a Phase 2 trial of 42 patients with early-stage chronic lymphocytic leukemia, standardised EGCG dosing was associated with a sustained reduction in circulating leukemia cells in roughly a third of patients and a meaningful reduction in swollen lymph nodes in more than two-thirds, over six months — real reduction in measurable disease burden, though no specific pathway from the mechanisms above was confirmed as the cause in these patients.
Expansion Suppression
Research concerning proliferation, cell-cycle progression, and the capacity of lesions to add durable mass.
EGFR / HER-family signaling
EGCG has been reported to block activation of the EGFR/HER2 receptor pair across five colon-cancer cell lines in one study, preferentially slowing the cancer cells' growth relative to normal colon cells from the same tissue.[1]
PI3K–AKT–mTOR
EGCG-driven AMPK activation has been reported to suppress this central growth-survival pathway in colon cancer cells, with a separate study implicating the same pathway in EGCG's fat-synthesis-blocking effect described under Starve above.[16,17]
JAK/STAT (STAT3)
In pancreatic cancer cells, EGCG has been reported to block a growth-signalling protein (STAT3), reducing the cells' ability to migrate and invade and enhancing their experimental response to gemcitabine and the JAK inhibitor CP-690550 — specific agents tested in this experimental system, not a general chemotherapy-sensitisation claim.[9]
TGF-β / SMAD signaling
In thyroid cancer cells, EGCG has been reported to block this growth-signalling pathway directly — the same mechanism responsible for its anti-invasion effect described under Contain above.[21]
Wnt / β-catenin
EGCG has been reported to suppress this pathway in colon cancer cells through direct degradation of its central signalling protein (β-catenin), independent of two enzymes (GSK-3β, PP2A) that would normally be required — a distinct mechanism from, but consistent with, its stemness-blocking effect described under Contain above.[5]
Attrition Pressure
Research concerning cellular stress vulnerability and net tumour-cell attrition under sustained conditions.
ER stress & unfolded protein response (UPR)
EGCG has been reported to trigger a cell-stress cascade in colorectal cancer cells, activating markers associated with the unfolded protein response and leading toward programmed cell death.[20]
EGCG's Attack classification is built on two distinct programmed-cell-death mechanisms reported in colorectal cancer cells. One demonstrates a mechanistically confirmed combination effect — EGCG sensitising cells to a death-receptor ligand — not direct tumour killing by EGCG alone. The other occurs alongside disrupted fat metabolism in a matched cell-and-animal study, with tumour-tissue-level effects measured directly in the animal arm. No human oncology trial has tested either mechanism, and the concentration gap covered under Pharmacokinetics and Administration below applies throughout.
Direct Tumour-Directed Killing
Research concerning regulated tumour-cell death (apoptosis, ferroptosis, necroptosis).
Extrinsic apoptosis (death receptors)
EGCG combined with a death-receptor-activating molecule (TRAIL) has been reported to trigger programmed cell death in colon cancer cells through a specific surface receptor (DR5) — removing that receptor blocked the effect entirely, confirming the mechanism directly rather than by correlation alone.[7]
Intrinsic apoptosis (mitochondrial / Bcl-2)
EGCG has been reported to trigger mitochondrial-damage-driven cell death in colorectal cancer cells in vitro, at concentrations (IC50 500–800 µM) far above what oral dosing has been confirmed to achieve in human plasma — detailed under the Concentration Gap section below. In a matched xenograft study using the same colorectal cell line and a substantial animal dose (30–50 mg/kg), EGCG-treated tumours showed reduced free fatty acid levels, reduced ATP levels, and altered lipogenesis-pathway protein expression directly in tumour tissue, consistent with the same apoptotic mechanism observed in culture.[17]
EGCG's Protect classification covers two distinct kinds of evidence, and both are limited enough to keep this role at partial rather than active. One is chemoprevention evidence tied to cancer risk rather than treatment, which has no defined mechanism by nature and is detailed below rather than carrying a pathway card — and here the human record is genuinely mixed, not a clear positive. The other is mechanism-based evidence that EGCG supports the body's own tissue resilience, which does carry a real pathway card, set out further down — but it rests on a single animal model of induced injury, and that finding sits in direct, stated tension with a risk covered under Safety and Interactions below, a tension this page states plainly rather than resolving one way or the other. A tumour-burden reduction reported in a human blood-cancer trial is discussed under Weaken above rather than here, since reducing circulating cancer cells is tumour-directed activity, not host protection, even though the trial itself measured a human clinical outcome.
Oncology Host-Status
Chemoprevention — genuinely mixed, and the products tested were mixed green tea catechin preparations, not isolated pure EGCG, so results cannot be attributed to EGCG alone. A randomised proof-of-principle trial in men with high-grade prostatic intraepithelial neoplasia found significantly fewer prostate cancer diagnoses after one year of green tea catechins versus placebo.[24] A larger, later randomised trial found no significant reduction in its primary one-year prostate-cancer-incidence endpoint; a secondary, exploratory analysis combining cancer diagnoses with a related precursor finding favoured the treatment arm, but does not overturn the negative primary result.[25] Separately, in mice, EGCG reduced diet-induced colon precancerous lesions, associated with activation of a cellular antioxidant defence pathway (Nrf2) — a real animal-level chemoprevention signal.[12]
Hepatic Resilience & Clearance
Human and preclinical research on hepatic enzyme systems, bile-acid handling, and liver-related markers.
Experimental hepatic injury modulation (induced-injury model)
EGCG has been reported to reduce liver scarring, mitochondrial oxidative stress, and inflammation in a mouse model of induced liver injury — a real, animal-corroborated finding, not purely a cell-dish result, though this specific study did not measure detoxification or bile-mediated clearance capacity directly, and that broader claim is not made here.[8] This sits in tension with a finding covered under Safety and Interactions below: concentrated EGCG extract carries a separate, well-documented, dose-dependent risk of liver injury in humans, with substantial variability between individuals. These findings come from fundamentally different models, doses, and exposure conditions — an induced-injury mouse model on one side, human idiosyncratic and dose-related susceptibility on the other — and the animal study cannot be used to infer protection against human green-tea-extract hepatotoxicity. Both are stated as real and unresolved against each other, not smoothed into a tidy healthy-versus-diseased-liver split.
Block Seeding & Niche Formation
Research concerning formation of supportive pre-metastatic niches at distant sites.
EGCG has been reported to suppress blood-vessel growth signalling in colon cancer cells, and in a matched animal model reduced tumour volume, new-vessel density, and cancer-cell growth while increasing tumour and vessel-cell death.
Prevent Dormant Reactivation
Research concerning wake-up signalling and reactivation of dormant disseminated tumour cells.
Cancer stemness (CD44, ALDH, Nanog/Sox2)
EGCG has been reported to block the stem-like state colorectal cancer cells rely on to survive treatment and reactivate, and restoring the underlying growth pathway reversed this effect directly.
Prevent Tumour Cell Shedding
Research concerning invasion and escape from existing lesions (EMT and ECM breach).
In thyroid cancer cells, EGCG has been reported to block the cellular transition tumour cells use to detach and invade, directly reducing invasion and migration.
Lipid Axis Pressure
Research concerning membrane synthesis and lipid-driven signalling capacity.
EGCG has been reported to trigger tumour-cell death in colorectal cancer cells by shutting down the fat-synthesis machinery those cells depend on, confirmed in both cell and animal models.
Glucose Axis Pressure
Research concerning glycolytic ATP production and glycolytic intermediates used by cancer cells.
Aerobic glycolysis (Warburg effect)
EGCG has been reported to reduce the glucose-uptake machinery colon cancer cells rely on for fast energy production, through activation of an energy-sensing enzyme.
Amino Acid / Protein Access Pressure
Research concerning nitrogen availability, amino-acid access, and biomass synthesis.
EGCG has been reported to block an enzyme tumour cells depend on to process glutamine, sensitising glioblastoma cells specifically under glucose or Akt-pathway stress, and triggering cell death in a high-risk childhood cancer cell model when combined with a second compound.
Expansion Suppression
Research concerning proliferation, cell-cycle progression, and the capacity of lesions to add durable mass.
EGCG has been reported to suppress this central growth-survival pathway in colon cancer cells, with a separate study linking the same pathway to its fat-synthesis-blocking effect.
Attrition Pressure
Research concerning cellular stress vulnerability and net tumour-cell attrition under sustained conditions.
ER stress & unfolded protein response (UPR)
EGCG has been reported to trigger a cell-stress cascade in colorectal cancer cells, activating markers associated with programmed cell death.
Direct Tumour-Directed Killing
Research concerning regulated tumour-cell death (apoptosis, ferroptosis, necroptosis).
Intrinsic apoptosis (mitochondrial / Bcl-2)
EGCG has been reported to trigger mitochondrial-damage-driven cell death in colorectal cancer cells; a matched xenograft study measured reduced fatty acid and ATP levels directly in tumour tissue, consistent with the same mechanism.
Hepatic Resilience & Clearance
Human and preclinical research on hepatic enzyme systems, bile-acid handling, and liver-related markers.
Hepatic injury modulation
EGCG has been reported to reduce liver scarring, oxidative stress, and inflammation in an animal model of induced liver injury. These findings come from a different model and exposure condition than EGCG's own human hepatotoxicity risk, so one should not be read as evidence against the other.
Advertisement
03 — Pharmacokinetics
Pharmacokinetics and Administration
How EGCG moves through the body — and how quickly it's cleared — has more bearing on real-world use than the mechanistic evidence alone. This is the practical layer that shapes formulation, timing, and co-medication decisions.
Absorption
Oral systemic exposure is low and highly variable. Fasted dosing substantially increases absorption compared with food, but fasting also increases nausea and may raise hepatotoxicity risk — higher absorption should not be read as the preferable way to take concentrated extract.
Highest Plasma Peak Under High-Dose Conditions
Against the highest single-dose peak measured in humans (~7.4 µM, at a 1,600 mg experimental dose), mechanistic findings vary widely in how far they sit above it: EGFR and TRAIL/DR5 findings are roughly 3- to 6-fold higher, while the lipogenesis finding used concentrations 70- to 110-fold higher — not an equal gap across the board.
Clinical Dose Context
Human trials span 50 mg to 2,000 mg EGCG twice daily. One review proposed 704 mg/day and 338 mg/day as review-derived reference levels, not accepted regulatory limits; EFSA separately found risk signals at ≥800 mg/day but could not establish a universally safe supplement dose.
Formulation Effects
A chemically modified, investigational EGCG prodrug outperformed native EGCG on tumour growth in matched breast and prostate animal models. This is a research compound, not an enhanced retail EGCG formulation, and neither has dedicated human safety data.
Metabolism
Enzyme-kinetic experiments identified sulfation as the predominant conjugation pathway under tested conditions, with substantially greater activity than methylation or glucuronidation. Between-person variability in these clearance enzymes is substantial.
Co-Dosing Considerations
Bortezomib, a chemotherapy drug, should be avoided with concentrated EGCG pending oncology review — the two can chemically interact, and whether this matters at real-world doses is genuinely disputed in the research. Sunitinib also warrants caution, based on animal and laboratory evidence rather than a controlled human study.
Absorption
The biggest practical constraint on using EGCG orally is the same one that runs through much of this page: oral systemic exposure is low and highly variable, and how much reaches the bloodstream depends heavily on when it's taken. A randomised trial testing a standardised EGCG-rich extract at three doses found that fasted administration substantially increased systemic exposure compared with dosing alongside a light breakfast.[2] This creates a real tension rather than a simple recommendation: fasting raises exposure, but it also increases gastrointestinal intolerance and may increase hepatotoxicity risk (see Safety below), and regulatory and USP guidance favours taking concentrated extract with food. Higher absorption from fasting should not be read as the preferable way to take concentrated EGCG.
Gastrointestinal side effects, particularly nausea, increased at higher doses and specifically under fasting conditions in that same trial.[2] EGCG is also cleared quickly once absorbed — its elimination half-life sits in the range of a few hours, not days — detailed further under Onset and Washout below.
Highest Plasma Peak Under High-Dose Conditions
This is essential context for reading the cell-model findings in Evidence Summary and Pathway Interaction Profile above: several of those mechanisms were reported at concentrations their own authors describe as in the double-digit micromolar range — above the highest concentration oral dosing has been confirmed to reach in human plasma, shown below. That highest confirmed concentration itself came from a single, very high experimental dose under controlled conditions, not from ordinary retail use, and should not be read as a typical "achievable" plasma level.
| Study | Dose | Peak plasma concentration |
|---|---|---|
| Ullmann 2003, single ascending dose (highest tested) | 1,600 mg pure EGCG | ~7.4 µM equivalent — the highest single-dose concentration measured in humans, under experimental conditions not representative of ordinary use[18] |
| Chow 2005, single dose | 1,200 mg Polyphenon E | Fasted dosing substantially increased exposure versus fed at every dose tested[2] |
| Scholl 2018, steady state (BID × 5 days) | 150 mg twice daily | Total exposure (AUC) ~878 µg·h/L, with more than 6-fold variability between individuals[33] |
Between-person variability is a real complication on top of the concentration gap itself: the same standardised dose produced more than a 6-fold spread in peak plasma levels across individuals in one steady-state trial.[33]
| Mechanism | Concentration used | Relative to human plasma peak |
|---|---|---|
| EGFR/HER2 inhibition (Shimizu 2005) | Tested 0–50 µM; IC50 ~44 µM | ~6-fold above[1] |
| TRAIL/DR5 sensitisation (Kwon 2020) | 20–40 µM | ~3- to 5-fold above[7] |
| Wnt/β-catenin suppression (Oh 2014, and colorectal EGCG studies generally) | Colorectal cell studies of this kind commonly span 0–150 µM | Broad range, mostly well above[5] |
| Lipogenesis/apoptosis (Khiewkamrop 2022) | IC50 500 µM (HCT116), 800 µM (HT-29) | ~70- to 110-fold above[17] |
These are not equally distant from what's achievable. The EGFR and TRAIL/DR5 findings sit within roughly an order of magnitude of the highest confirmed human plasma peak — a real gap, but a narrower one. The lipogenesis finding, by contrast — the same study underlying this profile's Starve and Attack animal corroboration — used concentrations one to two orders of magnitude higher, a substantially wider translational gap that the general "double-digit micromolar" framing elsewhere on this page doesn't fully convey. This is exactly why the specific numbers are reported here rather than a single blanket characterisation.
Clinical Dose Context
| Context | Dose | Source |
|---|---|---|
| Single ascending dose PK (Ullmann 2003) | 50–1,600 mg pure EGCG | Phase 1 PK/safety trial[18] |
| Single dose, fasted vs. fed (Chow 2005) | 400/800/1,200 mg Polyphenon E | Randomised PK trial[2] |
| Steady state (Scholl 2018) | 150 mg twice daily × 5 days | Population PK trial[33] |
| Early-stage CLL, Phase 2 (Shanafelt 2013) | Polyphenon E, standardised to 2,000 mg EGCG per dose, twice daily | Real biological-activity signal against measurable disease burden[4] |
| Prostate chemoprevention (Bettuzzi 2006) | ~300 mg EGCG/day (mixed catechins) | Positive on primary endpoint[24] |
| Prostate chemoprevention (Kumar 2015) | 400 mg EGCG/day (Polyphenon E) | Negative on primary endpoint[25] |
A systematic toxicology review proposed Observed Safe Levels of 704 mg EGCG/day for beverage-form extract and 338 mg EGCG/day for concentrated solid-bolus EGCG — review-derived estimates, not universally accepted regulatory upper limits.[10] The European Food Safety Authority reached a more cautious conclusion, identifying transaminase elevation at or above 800 mg/day in supplement form but stating that the available evidence was insufficient to establish a universally safe supplement dose, noting liver injury has occurred with products supplying as little as roughly 375 mg/day.[28] The CLL trial's standardised EGCG exposure above (4,000 mg/day total, split) exceeds these figures substantially — see Safety below for the adverse-effect data associated with that regimen.
Formulation Effects
EGCG's poor native bioavailability has prompted formulation research distinct from the standard-extract dosing described above. A chemically modified, investigational peracetate prodrug of EGCG — related peracetylated preparations across the two studies below, not two fundamentally different formulation strategies — designed to resist the compound's own rapid breakdown until it's inside a cell, produced significantly greater tumour-growth inhibition than native EGCG in a matched breast-cancer xenograft model, alongside greater proteasome inhibition and cell death in the same cells.[26] A related formulation reduced tumour growth and new blood-vessel formation while increasing tumour-cell death in an androgen-independent prostate-cancer xenograft model, compared against solvent-only controls.[27]
This is a chemically modified investigational compound, not an enhanced retail EGCG formulation in the usual nutraceutical sense. Neither study has dedicated human safety or pharmacokinetic data — both are animal-model findings, and the retail-extract absorption and safety data described elsewhere on this page do not necessarily transfer to this altered-kinetics form.
Metabolism and Interindividual Variability
EGCG undergoes extensive phase I and phase II metabolism, and one enzyme family dominates that process more than earlier general accounts suggested. A human ingestion study paired with direct enzyme-kinetic analysis found sulfation to be the primary phase II pathway under the conditions tested — roughly 2-fold more active than methylation and 60- to 300-fold more active than glucuronidation in human liver and intestinal tissue — carried out mainly by SULT1A1 in the liver and SULT1A3 in the intestine. These are enzyme-kinetic assay comparisons, not a direct measurement of total whole-body clearance. The major circulating human metabolite this produces, EGCG-4″-sulfate, reached plasma levels essentially equal to free EGCG itself.[31]
Methylation is carried out mainly by catechol-O-methyltransferase (COMT), and glucuronidation by UDP-glucuronosyltransferases — principally UGT1A1, UGT1A8, and UGT1A9 — both real, secondary pathways relative to sulfation under the tested conditions.[31,33] Genetic variation in these clearance enzymes is a plausible contributor to the more-than-6-fold between-person variability in plasma EGCG levels reported at steady state, but no specific genotype-exposure association has been established for EGCG the way it has for some other compounds on this site — this remains speculative.[33]
Co-Dosing Considerations
Two interactions have concerning mechanistic and preclinical evidence, despite the absence of a controlled human interaction trial for either. Each row is flagged by the most cautious guidance its cited evidence supports.
Discuss whether to combine, separate, or avoid EGCG and a medication with your treating oncology team or physician.
| Flag | Interaction |
|---|---|
| Avoid | Bortezomib and other boronic-acid-based proteasome inhibitors — EGCG's catechol and galloyl groups can form a direct chemical bond with bortezomib's boronic acid group; this antagonised bortezomib's anticancer activity in cell and xenograft models, with a companion pharmacology commentary recommending patients on bortezomib avoid green tea products.[11] Whether this happens at concentrations a person would realistically reach is genuinely disputed in the literature, not resolved: a follow-up xenograft study found antagonism only above 200 µM EGCG, with no antagonism at 11–16 µM — several-fold higher than the roughly 0.6–3 µM measured in humans taking EGCG supplements — leading those authors to conclude ordinary dietary or supplemental intake was unlikely to interfere with bortezomib.[13] A more recent study, however, reported that EGCG and related green tea catechins antagonised bortezomib at concentrations attainable through ordinary tea consumption under its experimental conditions.[34] Given that disagreement, and given how much is at stake if bortezomib's efficacy were compromised, concentrated EGCG and green tea extract products should be avoided unless the treating oncology team explicitly approves them — a conservative policy stated as a policy choice under scientific uncertainty, not as proven clinical treatment failure. |
| Caution | Sunitinib — a laboratory and rat study found that EGCG interacted physically with sunitinib, forming a precipitate and reducing sunitinib's plasma levels in the animal model; the investigation was prompted by a single clinical observation (tea consumption appearing to interfere with sunitinib symptom control in one patient with metastatic kidney cancer), not a controlled human interaction study.[32] Until human interaction data exist, concentrated EGCG use during sunitinib therapy should be reviewed with the treating oncology team. |
Advertisement
04 — Onset & Washout
Onset and Washout
EGCG is cleared from the body quickly — its plasma half-life is measured in hours, not days. What that fast clearance does and doesn't mean matters for interpreting the research, and for discussing timing with a clinician if you are also managing other medications.
Immediate Onset
Direct pharmacokinetic presence — peak plasma levels and clearance — is fast and well characterised. Whether this tracks the onset of any downstream pathway effect is a separate question, addressed below.
Steady State
Given a plasma half-life of roughly 2 to 4.6 hours, steady state on a twice-daily regimen is reached quickly by calculation. This applies to plasma concentration under a consistent regimen, not to downstream tumour-pathway effects.
Accumulated Effect
The human trials and most animal studies on this page used repeated dosing over weeks or months. Cell-model mechanisms more often followed a single, finite laboratory exposure. No study has established how many days of human dosing a downstream effect needs before it first appears.
Dosing Pattern in Studies
Human and animal outcomes on this page were studied under repeated dosing, not single doses. A continuity-dosing pattern is a reasonable inference from that and from EGCG's short plasma half-life, but no study has directly compared continuous against pulsed dosing for any of these outcomes.
Washout
How long EGCG's influence can take to clear before it stops being a relevant factor.
Earliest point at which most circulating parent EGCG is gone, calculated from its elimination half-life. This is a plasma-clearance estimate, not a clinically validated interaction-free window for drug interactions, surgery, or hepatotoxicity — those questions require medication-specific clinical guidance.
No human study identified here established a tissue-based washout interval for EGCG. Given that gap — and given that EGCG's hepatic and enzyme-interaction risks operate through a different mechanism than plasma presence alone — a cautious margin beyond plasma clearance is reasonable before surgery, a procedure, or a new medication with a narrow safety margin.
Two Distinct Clocks
EGCG produces two kinds of effect worth distinguishing — though only one of them has been precisely timed. Collapsing them into a single "does it work, and when" question is the most likely way this compound gets misread.
The direct-pharmacology clock (Clock A) is fast and well characterised: two independent human trials place peak plasma levels at roughly 1.3 to 2.2 hours after a dose, clearing again within a matter of hours given an elimination half-life of 1.9 to 4.6 hours.[18,19] This is the plasma presence available to directly engage the receptor- and enzyme-level mechanisms described in Pathway Interaction Profile above at any given moment.
| Clock A — Direct Pharmacology | Clock B — Downstream Phenotype | |
|---|---|---|
| Latency | Fast — within 1 to 2 hours[18,19] | Not established by a dedicated onset-timing study. The human trials and most animal studies on this page used repeated dosing over weeks or months[4,24,25], but several cell-model mechanisms in Pathway Interaction Profile above followed a single, shorter laboratory exposure — the two evidence types shouldn't be treated as requiring the same latency. |
| Persistence | Short — clears within hours, half-life 1.9 to 4.6 hours[18,19] | Not established. No study has measured how long any pathway-level or clinical-outcome effect persists after EGCG is stopped. |
| What it covers | Plasma presence — the circulating concentration available to engage the mechanisms described above | The pathway-level and clinical-outcome effects described throughout this page. Human and most animal outcomes required repeated dosing to appear; cell-model mechanisms more often reflect a finite exposure. The specific timing of onset and offset hasn't been separately measured for EGCG in either case. |
That gap in Clock B is worth stating plainly rather than papering over: the human clinical trials and most animal studies described on this page — the CLL trial, the chemoprevention trials, the animal chemoprevention and lipogenesis studies — used repeated, sustained dosing, not a single dose. That's real evidence that Clock B's clinical- and animal-level outcomes required continuity to appear. It is not evidence that every cell-model mechanism described in Pathway Interaction Profile required the same — several of those followed a single, finite exposure in culture, over hours to days rather than months. No study has directly compared continuous against pulsed human dosing for any oncology-relevant outcome on this page.
Steady State and Accumulation
With a half-life of roughly 2 to 4.6 hours, standard pharmacokinetic principles place EGCG at steady state within less than a day of consistent twice-daily dosing — this is a calculated inference from the half-life data above, not a separately measured steady-state timeline, but it is a standard and reliable extrapolation. The trial that most directly measured multi-dose EGCG levels (twice-daily dosing over five days) found more than a 6-fold difference in peak plasma concentration between individuals on an identical regimen.[33] That trial measured repeated-dose exposure and its substantial between-person variability — it did not report a first-dose-to-steady-state accumulation ratio, so the degree of true accumulation from twice-daily dosing shouldn't be assumed beyond what was actually reported.
Dosing Pattern in Studies
The underlying tension here is the same one that runs through this entire profile: the mechanistic case for EGCG is broad, spanning all five functional roles described in Pathway Interaction Profile above, but the gap between what's mechanistically plausible and what's clinically achievable — detailed in the Concentration Gap section under Pharmacokinetics and Administration above — limits how much real-world confidence that mechanism deserves. That gap, not any single weak result, is what keeps EGCG's oncology-relevant case broad but preliminary rather than clinically established.
EGCG is provisionally best described as studied primarily under repeated dosing, not as a demonstrated continuity requirement. Clock A clears within hours, and the human and most animal outcomes described above came from sustained dosing rather than a single dose — but a short plasma half-life doesn't by itself prove continuous dosing is superior to a pulsed regimen, since downstream effects may outlast plasma EGCG and no pulse-versus-continuous comparison exists for any outcome on this page. Single-day or occasional use remains the less-studied option — see Clinical Dose Context under Pharmacokinetics and Administration above for what dosing regimens have actually been studied.
Washout
EGCG's plasma clearance itself is well characterised — most circulating parent EGCG is gone within roughly a day of the last dose, calculated directly from the elimination half-lives described above.[18,19,33] This describes parent-compound plasma presence specifically; some metabolites and downstream pathway effects may persist longer than that estimate implies. What is not established is a separate, dedicated washout study measuring how long EGCG's downstream pathway effects — as opposed to its plasma presence — persist after stopping. No human study identified here established a tissue-based washout interval for EGCG.
Given that gap, plasma clearance (roughly a day) is a reasonable floor for thinking about washout, not a validated ceiling or a generic margin sufficient for any specific purpose — particularly given EGCG's hepatic and enzyme-mediated interaction risks, detailed under Safety and Interactions below, which are governed by a different mechanism than plasma presence alone. No clinically validated washout interval exists for drug interactions, surgery, or hepatotoxicity specifically; those questions require medication-specific clinical guidance rather than a general margin. A cautious approach is to raise EGCG use with a care team as soon as it begins rather than waiting for any fixed window to close.
Advertisement
05 — Safety
Safety Profile
EGCG is generally well tolerated at moderate doses, but its best-characterised risk — dose-dependent liver injury from concentrated extract — carries real weight in deciding how it should be used, and its drug interactions matter specifically in an oncology context.
Transaminitis — liver enzyme elevation reported in about a third of patients in a monitored Phase I oncology trial, all grade 1.
Nausea and abdominal pain — the most common tolerability effects, increasing at higher doses and specifically under fasting.
Hepatotoxicity — concentrated extract carries a real liver-injury risk with individual variability; regulators have required food-based dosing and cautionary labelling, though a universally safe supplement dose has not been established.
Drug interactions — concentrated EGCG has shown preclinical antagonism with the chemotherapy drug bortezomib and a preclinical interaction with sunitinib; oncology team review is warranted before combining with either.
Adverse Effects in Human Trials
The clearest safety signal from human trials is that EGCG's most common adverse effects are mild-to-moderate, not severe. In the Phase I dose-finding trial for Polyphenon E standardised to 2,000 mg EGCG per dose, twice daily — the same regimen later used in the Phase 2 CLL trial described in Evidence Summary above — the maximum tolerated dose was not reached at that highest dose level tested.[14] That finding describes a monitored oncology trial in a selected patient population under clinical supervision, not evidence that this dose is broadly safe for unsupervised self-administration. Transaminitis (liver enzyme elevation) occurred in about a third of patients, all grade 1. Abdominal pain occurred in roughly three in ten patients at grade 1, with a small proportion at grade 3. Nausea occurred in around four in ten patients at grade 1, with a smaller share at grade 2.[14] Separately, gastrointestinal effects and headache increased at higher doses and specifically under fasting conditions in single-dose PK trials.[2,18]
Hepatotoxicity
Liver injury from concentrated extract is EGCG's single most clinically significant adverse signal, and the one most directly relevant to how it should be dosed. The relationship is real but not a simple linear dose curve: risk rises with higher supplement exposure, particularly around or above 800 mg/day, but green-tea-extract hepatotoxicity also shows substantial variability between individuals, product-composition effects, and occasional cases reported at considerably lower doses — features more consistent with idiosyncratic, susceptibility-dependent injury layered on top of a genuine dose relationship, not a single clean threshold.[28,29] A systematic toxicology review proposed Observed Safe Levels of 704 mg EGCG/day for beverage-form extract and 338 mg EGCG/day for concentrated solid-bolus EGCG — review-derived estimates, not universally accepted regulatory upper limits.[29] The European Food Safety Authority reached a more cautious conclusion: it identified transaminase elevation at or above 800 mg/day in supplement form but stated that the available evidence was insufficient to establish a universally safe supplement dose, noting that liver injury has occurred with products supplying as little as roughly 375 mg/day.[28]
This risk has prompted real regulatory action, and different authorities have taken different, non-interchangeable approaches worth distinguishing rather than merging into one global recommendation. The U.S. FDA temporarily suspended human oral EGCG trials in 2006 pending additional animal toxicity review; trials were later permitted but required to administer EGCG with food.[28] The United States Pharmacopeia, following its own comprehensive hepatotoxicity review, added a cautionary labelling requirement to its green tea extract monograph: do not take on an empty stomach, take with food.[29] Health Canada separately mandated a liver-injury warning label on green tea extract products, including guidance to consult a healthcare practitioner before use in the presence of a pre-existing liver disorder.[30]
This risk sits in tension with a separate finding described under Pathway Interaction Profile above: in a mouse model of induced liver injury, EGCG reduced liver scarring, oxidative stress, and inflammation.[8] Both facts are real, but the relationship between them is not fully resolved rather than neatly split — human green-tea-extract liver injury is often idiosyncratic with variable individual susceptibility, and a bile-duct-ligation mouse model does not establish protection in human liver disease. The animal finding cannot be used to infer protection against human hepatotoxicity risk. The CLL trial's standardised EGCG exposure (4,000 mg/day total, split) exceeds the safe-level estimates above substantially, and may have contributed to the transaminase elevations reported in that trial's dose-finding predecessor — though the trial did not establish a specific dose-response causal mechanism, and population, product composition, treatment duration, and fasting status could all have contributed as well.[10,14]
A note on product identity: the hepatotoxicity literature spans pure EGCG, standardised extracts such as Polyphenon E, mixed green tea catechin preparations, and brewed tea — these are not interchangeable exposures. Total catechin content, extraction method, product matrix, and bolus-versus-beverage or fasted-versus-fed administration can all alter risk independent of the EGCG dose alone.
06 — Sourcing
Sourcing Guide
Formulation is a major factor in whether an EGCG product can deliver anything close to what the research above describes — and, given the hepatotoxicity risk tied specifically to concentrated bolus dosing, dosing format matters here more than for some other compounds on this site. Our Sourcing Guide offers a curated list of products available on the retail market we found to answer these concerns.
EGCG Sourcing Guide07 — Literature
References
Last reviewed: July 2026